Master Browser errors 1

[BWayne2k]

Can anyone help me out with this one, the situation is as follows: The NT Server 4 that I have is giving me the following error
"Master Browser has received a server announcement from computer (wrkstn name) that believes it is the master browser for domain on Transport Nbf_E100B1. The master browser is stopping or an election is being forced."

I have read the forums that have been posted here and in some other places where they say that it might be the Registry key, I have checked that and everything is ok. I have also checked that the computers have the correct subnet mask (192.162.1.* network) (255.255.255.0 subnet)
Nothing that I have found has worked, can anyone give me some other ideas. The main problem here is that when that message appears on the server, the whole network goes down. It probably happens maybe twice a day, and there is no specific time. So let me know, thanks

 

[ShackDaddy]

I think that the error you are seeing is a symptom of the problem, not the cause. Normally these errors have nothing to do with causing a 'network to go down' but may be caused by the inability of another system to contact your NT4 server. If another system fails to contact your NT4 server, it will throw a browser election and assume the master browser role. When your server wakes up and realizes that there is another master browser, it fires off this message. The troubling question is where the heck the server was when the workstation elected itself master browser.

I tend to think that your network goes down before the message appears and only appears once the network is on its way to recovery. What else is going on in the event logs?

You might pick one of your systems, open a command-line, set the vertical screen buffer at 200 lines, and do a 'ping servername -t'. Just leave the ping running. Next time your system seems to 'go down' take a look at whether the client stopped being able to ping the server. That should give you a general idea of the length of the outage and/or whether the outage was service based (problems with file sharing or SMB) or IP based, in which case you'd see trouble with pings.

ShackDaddy

 

[BWayne2k]

The only other message that I get is: Is_jkserveruser (the jkserver part is the servername) was unable to log on, bad username or password
That message, together with the browser error from like, about 6 or seven different Win98 machines

 

[ShackDaddy]

This error means that the Anonymous user account in IIS on your server is probably misconfigured. It shouldn't have anything to do with the loss of network connectivity or your browsing errors.

 

[rlee1]

You have a Browser War going on. No big deal, it just adds overhead to network traffic. One machine wants to "hold" the browser list that you see in the Network Neighborhood Do a search in Technet on "Master Browser" and you can find info.

also see

http://www.microsoft.com/technet/tr...TechNet/prodtechnol/ntwrkstn/tips/ewn9662.asp

and

http://www.microsoft.com/technet/tr...prodtechnol/winntas/tips/platinum/ptbimsn.asp

 

[wimvanherp]

I have had about the same problem and it was because I had a NT-cluster and the inter-clusterlink had an ip-adress not complete out of the range of the normal adresses.
You can also prohibit the workstations to be a master browser (in the registry) but probably you have checked this already.. Wim Vanherp
Wim.Vanherp@belgacom.net

 

[ChrisRAnthony]

Using the registry editor open HKLM\system\currentcontrolset\services and set the key MaintainServerList to NO. When this is set to NO the system does not cache the list of browser announcements it hears, so it cannot become a browse server. When this value is set to AUTO the list is cached and the computer may force an election. Whne it is set to YES, the computer always acts as a browse server.

 

[ShackDaddy]

Napegator, when you say that the network goes down, what do you mean? Do you mean that nothing shows up in Network Neighborhood, or that drives mapped to the server no longer work? If it's just that you are missing network neighborhood, that would be browse-related. If mapped drives don't work, that's something else completely.

ShackDaddy

 

[BWayne2k]

The network goes down in the sense that some computer are able to see the server, and some are not. I checked the hub and it seems to be working, I just can't ping the server by ip or name. There are about 35 workstations and the when the problem happens, only about 5 or six pcs can see the server, eventhough there seems to be a connection active (there is a light on the network card of the pcs and the hubs) but I can't connect to the server

 

[erimar]

I would start by turning off the computer browser service under services in the control panel of all the workstations. This cuts down on traffic and then I would see what happens as far as the "Going Down" situation. All the pc's can still see the network they will just look at the Server for the database. This would also stop your "Master Browser has received a server announcement from computer (wrkstn name) that believes it is the master browser for domain on Transport Nbf_E100B1. The master browser is stopping or an election is being forced."

 

[BWayne2k]

Actually, I spoke with a friend of mine, he told me that he was having the same problem and what he had to do was replace the hub. I ordered a hub and just go it in last night. I replaced it (we have 3/24port hubs,I just replaced the bottom one with the new one) and the network seems to be up and running so far.(Since last night it hasn't gone down) It would usually go down in the morning and in the afternoon, so I'll see how everything works.

 

[dasein47]

I've had similar problems before and the best way to deal with it, if this is possible, is to unplug the server from the network and reboot it. and if you can unplug all the connected workstations from the hub, do that too. make sure the server is the only one rebooting on the hub-network, let it come back up, it will force itself to be the master browser. then plug all the other workstations back in and you should be good to go.

i only suggest this cause i had the same problem in a network of about 10 PCs. i know this may not be possible or easy, but it fixes it quickly/minimizes downtime.

kip